Trust & Factuality · sub-raceNarrow lead
Factuality race
Answering knowledge questions correctly, or abstaining instead of inventing.
- Leader
- Fable 5.1
- Margin
- 0.8 pts
- 0.8% ahead
- Weeks at #1
- 9
- in a row
- Changes
- 5
- in 2026
Warming up the track…
Oct 19, 2026Fable 5.1 leads
2025Apr ’25Jul ’25Oct ’252026Apr ’26Jul ’26Oct ’26
Score history
Top 6 today over the leader's frontier · click to jump to a week
#1 frontierFable 5.1Opus 5.5Gemini 4 ArgonGPT-6 AstraGPT-6.1 SolSonnet 5.5
Reigns at #1
Longest: Opus 4.1, 13 weeks (from Aug 2025).
Sibling races
More trust & factuality races
Why this race matters
Hallucinations are the number-one reason people distrust AI answers.
How it is scored
Accuracy minus hallucination penalty, rescaled.
Unit: percentHigher is betterPhoto finish under 0.5 ptsDominant over 3.0 pts
Data provenance
Rewards abstention over confident errors.
Scores here are illustrative placeholders that mimic real benchmark patterns. Swap in live data through the score provider. Methodology